Henderson, L. (2007). Invasive, Naturalized and Casual Alien Plants in Southern Africa
Total Page:16
File Type:pdf, Size:1020Kb
Bothalia 37,2: 215–248 (2007) Invasive, naturalized and casual alien plants in southern Africa: a sum- mary based on the Southern African Plant Invaders Atlas (SAPIA) L. HENDERSON* Keywords: biomes, casual alien plants, invasive plants, Lesotho, naturalized plants, roadside surveys, SAPIA mapping project, South Africa, Swaziland ABSTRACT The primary objective of this publication is to provide an overview of the species identity, invasion status, geographical extent, and abundance of alien plants in South Africa, Swaziland and Lesotho, based on fi eld records from 1979 to the end of 2000. The dataset is all the species records for the study area in the Southern African Plant Invaders Atlas (SAPIA) database during this time period. A total of 548 naturalized and casual alien plant species were catalogued and invasion was recorded almost throughout the study area. Most invasion, in terms of both species numbers and total species abundance, was recorded along the southern, southwestern and eastern coastal belts and in the adjacent interior. This area includes the whole of the Fynbos and Forest Biomes, and the moister eastern parts of the Grassland and Savanna Biomes. This study reinforces previous studies that the Fynbos Biome is the most extensively invaded vegetation type in South Africa but it also shows that parts of Savanna and Grassland are as heavily invaded as parts of the Fynbos. The Fabaceae is prominent in all biomes and Acacia with 17 listed species, accounts for a very large proportion of all invasion. Acacia mearnsii was by far the most prominent invasive species in the study area, followed by A. saligna, Lantana camara, A. cyclops, Opuntia fi cus-indica, Solanum mauritianum, Populus alba/×canescens, Melia azedarach, A. dealbata and species of Prosopis. INTRODUCTION transects of between fi ve and 10 km long. Recordings of streambank species were made at virtually all water- History of roadside surveys in South Africa course crossings on the survey route. Roadside surveys of invasive plants in South Africa The Southern African Plant Invaders Atlas mapping were pioneered by Henderson and Musil (née Duggan) project (SAPIA) starting in 1979 in the central Transvaal, now Gauteng (Wells, Duggan & Henderson 1980), with the remainder The Southern African Plant Invaders Atlas (SAPIA) is of the Transvaal surveyed in 1982 and 1983 (Henderson a mapping project, launched in January 1994, to collate & Musil 1984). Surveys of the rest of South Africa were information on the distribution, abundance and habitat conducted by Henderson from 1986, starting with Natal types of invasive and naturalized alien plants in southern (Henderson 1989), followed by the Orange Free State Africa (Henderson 1998b). The fi rst phase of SAPIA, (Henderson 1991a), northern Cape (Henderson 1991b), involving volunteer participants, was scheduled for a eastern Cape (Henderson 1992), western and central fi ve-year period, ending in December 1998. The atlas Cape (completed in 1993 but unpublished), and southern region covered South Africa, Lesotho and Swaziland. and southwestern Cape (Henderson 1998a). Information was recorded on two standardized atlas All terminology used in this paper relating to invasive sheets, with slightly different species lists, covering the plants such as ‘alien’, ‘invasive’, ‘naturalized’, ‘casual western and eastern halves of the atlas region. One hun- alien’, ‘weed’ and ‘environmental weed’ conforms, as dred plant taxa were listed on each sheet, with a com- far as possible, to the defi nitions provided by Richardson bined total of 161 species. A pocket fi eld guide was com- et al. (2000) and Pyšek et al. (2004). The method used piled to help with the identifi cation of all listed species in these surveys was designed initially to make use of (Henderson 1995). otherwise unproductive travelling time whilst engaged in other research projects. The method was refi ned as SAPIA database the surveys progressed until a standardized method was developed (see Henderson 1992, 1998a). The presence A computerized SAPIA database was created by and abundance of all alien trees, large shrubs and con- incorporating all Henderson survey data (± 23 000 spicuous climbers which appeared to be naturalized or records) and SAPIA phase one project data (± 20 000 occurring outside of cultivation were recorded for each records). The SAPIA project continued on an ad hoc veld type category, habitat type (roadsides and adjoining basis and by the end of 2000 a total of ± 48 000 records veld, and streambanks) and quarter-degree/fi fteen minute had been accumulated. Thereafter, the SAPIA initiative square traversed by road. dwindled due to lack of funding. Only 10 000 records were added in the fi ve year period from 2001 to the Recordings of species on roadsides and in the adja- end of 2005. The SAPIA project was revived in 2006 cent veld were made from a moving vehicle along road with funding from the Department of Water Affairs and Forestry’s Working for Water Programme. The SAPIA * Agricultural Research Council: Plant Protection Research Institute, c/o SANBI, Private Bag X101, 0001 Pretoria. database has been computerized using Microsoft Access E-mail: [email protected] and is housed at the Plant Protection Research Institute MS. received: 2006-10-06. in Pretoria. 216 Bothalia 37,2 (2007) Objectives of this study Data treatment • To provide an overview of the species identity, inva- Abundance sion status, geographical extent, and abundance of alien plants in South Africa, Swaziland and Lesotho, Species abundance ratings in the SAPIA database are based on fi eld records from 1979 to the end of 2000. qualitative estimates. Table 1 shows the abundance rat- ings used in the SAPIA database and the equivalent rat- • To highlight the most prominent invaders in the ing used in Henderson surveys. For the purposes of this region as a whole, in each of the biomes, and in study, species abundance ratings were converted to a riparian and wetland habitats. numerical value as done in previous surveys (Henderson • To compare invasion and provide species profi les for 1998a) and each abundance rating was expressed in each of the biomes. numbers of individuals or groups per 10 km transect/ recording (Table 1). METHODS Prominence A similar formula was used in this study to calcu- Sampling method late prominence as in previous studies by Henderson The dataset for this study is all the species records for (1998a). The prominence value of a species x in category South Africa, Swaziland and Lesotho in the SAPIA data- y (biome or study area) was calculated as follows: base collected from 1979 until the end of 2000 (± 48 000 total abundance of species x in category y records). During this period a concerted effort was made × 100 to gather as much data from as wide an area as possi- sum of the abundances of all species in category y ble. The information gathered is the best available data prominence = + concerning the extent of invasion and species composi- value tion, at least of the larger trees, shrubs and conspicuous total species records of species x in category y climbers, in the study area over this time period. × 100 sum of the records of all species in category y The SAPIA dataset was subdivided on a quarter- degree square (QDS) basis into six datasets representing The highest prominence values in a given category the biomes of southern Africa. According to Rutherford which add up to ± 160 points out of a total of 200 are (1997) there are seven biomes in southern Africa: Savanna, printed in bold in Appendices 1–3. The cut-off point is Fynbos, Forest, Grassland, Nama-Karoo, Succulent Karoo arbitrary but represents the upper 80% of the summed and Desert. The Forest Biome in southern Africa is min- prominence values. iscule, only occurring in the Knysna area. However, if all the forest patches elsewhere are included, its area increases several-fold (Rutherford 1997). In this study RESULTS Forest refers to the Forest Biome and also forest habitats within the Savanna, Fynbos and Grassland Biomes. The A total of 548 naturalized and casual alien plant spe- Desert Biome occurs almost exclusively in Namibia, cies were catalogued in the SAPIA database for South except for a very small patch along the Orange River Africa, Swaziland and Lesotho from 1979 to the end of bordering on South Africa that has been excluded from 2000 (Appendix 4). At least 119, mainly herbaceous, this study. taxa are considered to have been under-recorded and TABLE 1.—Abundance ratings used in Henderson surveys, SAPIA and this study Henderson surveys SAPIA This study Rating Roadsides and veld No.* Streambanks Rating All habitats† All habitats# 9 A virtually continuous, almost pure stand 1 000+ Any number, with cover more than 7 Very abundant 1 000 75% of the reference area 8 The commonest species in a generally con- 500–999 Any number, with 50–75% cover 6 Very abundant 1 000 tinuous tree or shrub layer 7 Less abundant than above but > 20 individu- 200–499 Any number, with 25–50% cover 5 Abundant 200 als or groups per km 6 10–20 individuals or groups per km 100–199 Any number with 5–25% cover 4 Abundant 200 5 5–10 individuals or groups per km 50–99 Numerous, but < 5% cover or scat- 3 Frequent 50 tered, with cover up to 5% 4 2–5 individuals or groups per km 20–49 Few, with small cover 2 Frequent 50 3 ± 1 individual or group per km 5–19 Solitary, with small cover 1 Occasional 10 2 Less abundant than above but more than 1 2–4 Occasional 10 individual or group per 5 km 1 ± 1 plant or group per 5–10 km 1 Rare 1 *, approximate numbers of individuals or groups per 10 km transect.